The Innovation Theory of Harm: An Appraisal

• Author: Vincenzo Denicoló and Michele Polo
• Position: Professor of Economics, University of Bologna, and CEPR/Professor of Economics, Bocconi University, and GREEN (Center for Geography, Resources, Environment, Energy and Network)
• Pages: 921-953

THE INNOVATION THEORY OF HARM:

AN APPRAISAL

V

INCENZO

D

ENICOL `

O

M

ICHELE

P

OLO

*

There is a broad consensus that horizontal mergers may increase market

power but may also bring cost savings. Antitrust authorities and the courts

routinely seek to maximize allocative efficiency by assessing these opposing

effects. The details of this exercise may vary across jurisdictions or individual

cases, but on the whole the short-run questions involved in horizontal merger

control would appear to be largely settled.

Static allocative efficiency, however, is just part of the story. Antitrust au-

thorities are also concerned that horizontal mergers may affect innovation,

and hence dynamic efficiency.

1

When it comes to dynamic effects, the assess-

ment of mergers is on much less solid ground. The relationship between com-

petition and innovation has been explored by a vast literature, both theoretical

and empirical, but results remain ambiguous. Richard Gilbert concludes his

survey of the literature by noting that “[w]e remain far from a general theory

of innovation competition.”

2

In his view, the literature has shown that com-

petition may either increase or decrease innovation, depending on the cir-

cumstances. Significant progress has been made in identifying which

circumstances are relevant, but a broad consensus is still lacking.

3

* Vincenzo Denicol`o, Professor of Economics, University of Bologna, and CEPR. Michele

Polo, Professor of Economics, Bocconi University, and GREEN (Center for Geography, Re-

sources, Environment, Energy and Network). We thank, without implicating, Giacomo Calzolari,

Giulio Federico, Gregor Langus, Massimo Motta, Jorge Padilla, Emanuele Tarantino, Tommaso

Valletti, Piercarlo Zanchettin, and participants at the 2018 Annual MaCCI conference (Mann-

heim) for useful comments and discussion. Financial support from Compass Lexecon and Baker

McKenzie is gratefully acknowledged.

1

For example, Richard J. Gilbert and Hillary Greene report that since 2004, U.S. agencies

mentioned innovation effects in over one-third of the mergers challenged. Richard J. Gilbert &

Hillary Greene, Merging Innovation into Antitrust Agency Enforcement of the Clayton Act, 83

G

EO

. W

ASH

. L. R

EV

. 1919, 1932–33 (2015).

2

Richard Gilbert, Looking for Mr. Schumpeter: Where Are We in the Competition-Innovation

Debate? in 6 I

NNOVATION

P

OLICY AND THE

E

CONOMY

159, 206 (Adam B. Jaffe et al. eds., 2006).

3

Id. at 204–06.

921

82 Antitrust Law Journal No. 3 (2019). Copyright 2019 American Bar Association. Reproduced

by permission. All rights reserved. This information or any portion thereof may not be copied

or disseminated in any form or by any means or downloaded or stored in an electronic

database or retrieval system without the express written consent of the American Bar

Association.

922 A

NTITRUST

L

AW

J

OURNAL

[Vol. 82

In light of this, antitrust authorities have generally taken a cautious ap-

proach, limiting intervention mostly to cases in which the merging firms’ in-

novative products are close to commercialization (the “product pipeline”). In

these cases, innovation outcomes have been regarded as predictable enough to

be amenable to the standard, static analysis. The traditional tools adopted to

analyze the impact of the merger on competition among existing products

have then been applied also to those new products that would be marketed in

the near future.

But policy now seems to be changing, especially in Europe. In a series of

decisions that culminated in Dow/DuPont, the European Commission has

gradually shifted the focus of its dynamic merger analysis from product pipe-

lines to “innovation markets.”

4

According to the Commission, the effect of

mergers on such innovation markets is generally undesirable. In the Dow/

DuPont decision, for instance, the Commission writes:

The merger between [two firms] will result in internalization by each merg-

ing party of the adverse effect of the R&D projects on . . . the other merging

party; hence . . . it will reduce investment in the competing R&D projects.

The innovation competition effect follows the basic logic of unilateral ef-

fects, which is equally applicable to product market competition and to inno-

vation competition.

5

Having articulated the view that horizontal mergers generally stifle innova-

tion, the Commission concludes that a merger may have a negative impact on

innovation, reducing R&D investment and slowing down technological pro-

gress, independently of the more traditional effects on product market compe-

tition.

6

Thus, even mergers for which the static effects are benign could be

regarded as anticompetitive in a dynamic perspective. In the policy debate,

this approach has come to be known as the innovation theory of harm (IToH).

This article offers an appraisal of the IToH from an economic perspective.

In particular, we critically discuss recent work by Giulio Federico, Gregor

4

Nicolas Petit documents this shift in emphasis through a detailed analysis of recent Euro-

pean merger cases. He describes the decision in the Dow/DuPont case as a “small but significant

change in merger policy.” Nicolas Petit, Innovation Competition, Unilateral Effects, and Merger

Control, supra this issue, 83 A

NTITRUST

L.J. 873, 877 (2019). Petit also points out that a similar

shift was considered by U.S. agencies in the past but seems to have been abandoned. Id.

5

Case M.7932—Dow/DuPont, Comm’n Decision ¶ 145 (Mar. 27, 2017) (summary at 2017

O.J. (C 353) 9), ec.europa.eu/competition/mergers/cases/decisions/m7932_13668_3.pdf.

6

In Dow/DuPont, the Commission makes a clear distinction between the product markets

where the two merging firms might have competed and the overlap between their research activi-

ties. See id. ¶¶ 303–332; Press Release, Eur. Comm’n, Mergers: Commission Clears Merger

between Dow and DuPont, Subject to Conditions (Mar. 27, 2017) (IP/17/772). The assessment of

these latter overlaps does not refer only to pipeline research projects, i.e., innovative products

close to the marketing phase, but extends to the merging firms’ research labs and long-term R&D

projects.

2019] T

HE

I

NNOVATION

T

HEORY OF

H

ARM

: A

N

A

PPRAISAL

923

Langus, and Tommaso Valletti

7

and Massimo Motta and Emanuele Tarantino

8

that are often regarded as providing the theoretical underpinnings of the IToH.

Both Federico et al. and Motta and Tarantino highlight conditions under

which horizontal mergers affect innovation adversely.

9

Both articles acknowl-

edge the apparent contradiction between their clear-cut conclusions and the

mixed findings of the literature on innovation competition.

10

They justify this

difference by noting that the general findings of the innovation competition

literature do not necessarily apply to horizontal mergers. What distinguishes

horizontal mergers is that they do not reduce competition uniformly but rather

increase the market power of a subset of the firms. This might explain why

the effect of horizontal mergers on innovation may be more negative than the

innovation competition literature suggests.

11

While these articles are interesting analytical contributions, they depend on

restrictive assumptions and thus do not fully account for possible innovation-

enhancing effects that can exist when competing firms merge. In this article,

we highlight those assumptions and show how relaxing the assumptions can

lead to conclusions at odds with the IToH. From this analysis, we conclude

that the economic underpinnings of the IToH are in fact too fragile to serve as

the foundation for any general presumption that horizontal mergers have

harmful effects on innovation.

Let us now briefly summarize our main arguments. In spite of the similarity

of their results, the mechanisms analyzed by Federico et al. and Motta and

Tarantino are in fact quite different.

Federico et al.’s analysis can in principle offer support to the IToH. They

consider a model where firms that invest in research may duplicate the same

innovation or produce innovations that are close substitutes. This creates spe-

7

See generally Giulio Federico et al., Horizontal Mergers and Product Innovation, 59 I

NT

’

L

J. I

NDUS

. O

RG

. 1 (2018) [hereinafter Federico et al., Horizontal Mergers and Product Innova-

tion]; Giulio Federico et al., A Simple Model of Mergers and Innovation, 157 E

CON

. L

ETTERS

136 (2017) [hereinafter Federico et al., Mergers and Innovation].

8

See generally Massimo Motta & Emanuele Tarantino, The Effects of Horizontal Mergers,

When Firms Compete in Prices and Investments (Universitat Pompeu Fabra, Dep’t Econ. and

Bus., Working Paper No. 1579, 2018), sites.google.com/site/massimomottawebpage/.

9

For example, Federico et al. summarize their results as follows: “We find that a merger

reduces the incentives to innovate for the merging parties, absent efficiencies or spillover effects

that would reduce appropriability ex post.” Federico et al., Mergers and Innovation, supra note 7,

at 139. Similarly, Motta and Tarantino write: “[U]nder no (or weak enough) efficiency savings a

merger will reduce aggregate investment and harm consumers.” Motta & Tarantino, supra note

8, at 2.

10

This literature analyzes how changes in the intensity of competition, as measured for in-

stance by variations in the number of firms or the degree of product differentiation, affects inno-

vative activity. See generally Gilbert, supra note 2 (for references).

11

See Federico et al., Horizontal Mergers and Product Innovation, supra note 7, at 3; Motta

& Tarantino, supra note 8, at 3–4.